Linking KANSAI and Image Features by Multi-layer Neural Networks

Authors:
Xinyin Huang;Shouta Sobue;Tomoki Kanda;Yen-Wei Chen
Affiliations:
School of Education, Soochow University, Suzhou, Jiangsu 215006, China;College of Information Science and Eng., Ritsumeikan Univ., Shiga, 525-8577, Japan;College of Information Science and Eng., Ritsumeikan Univ., Shiga, 525-8577, Japan;College of Information Science and Eng., Ritsumeikan Univ., Shiga, 525-8577, Japan
Venue:
KES '07 Knowledge-Based Intelligent Information and Engineering Systems and the XVII Italian Workshop on Neural Networks on Proceedings of the 11th International Conference
Year:
2007

Citing 2
Cited 0

Media Converter with Impression Preservation Using a Neuro-Genetic Approach

International Journal of Hybrid Intelligent Systems
An efficient color representation for image retrieval

IEEE Transactions on Image Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

KANSEI is a Japanese term which means psychological feeling or image of a product. KANSEI engineering refers to the translation of consumers' psychological feeling about a product into perceptual design elements. Recently several researches have been done for image indexing or image retrieval based on KANSEI factors. In this paper, we report a quantitative study on relationship between image color features and human KNASEI factors. We use the semantic differential (SD) method to extract the KANSEI factors (impressions) such as bright, warm from 4 group subjects (Children, students, adults, elderly person) while they viewing an image (painting). A neural network is used to learn the mapping functions (relationships) from the image feature space to human KANSEI factor space (psychological space).